Photo Credit: Wildpixel

The following is a summary of “Comparison in prostate cancer diagnosis with PSA 4–10 ng/mL: radiomics-based model VS. PI-RADS v2.1,” published in the October 2024 issue of Urology by Li et al. 

Researchers conducted a retrospective study to evaluate the accuracy of MRI-based radiomics in diagnosing prostate cancer (PCa) in patients with PSA levels between 4 and 10 ng/mL, comparing it to the Prostate Imaging Reporting and Data System (PI-RADS v2.1) score. 

They analyzed 221 patients with prostate lesions and PSA levels between 4 and 10 ng/mL, including 154 in the training group and 67 in the validation group. Pathological confirmation was obtained via MRI-TRUS fusion targeted biopsy or systematic transrectal ultrasound (TRUS) guided biopsy. From ADC and T2WI images, 851 radiomic features were extracted. The least absolute shrinkage and selection operator (LASSO) regression and logistic regression were used for feature selection and model development. The diagnostic accuracy of the radiomics models and PI-RADS v2.1 score was evaluated. 

 The results showed that 10 radiomic features were selected from ADC images, 13 from T2WI images, and 7 from combined models. The ADC, T2WI, and combined models achieved diagnostic accuracy in the training groups [AUC: 0.945 (ADC), 0.939 (T2WI), 0.979 (combined)] and validation groups [AUC: 0.942 (ADC), 0.943 (T2WI), 0.959 (combined)], significantly exceeding the PI-RADS v2.1 model [0.825 for training cohort and 0.853 for validation cohort]. The 3 radiomics models showed superior PCa diagnostic performance compared to PI-RADS v2.1 in both training (P < 0.001) and validation groups (P = 0.015). 

The study concluded that radiomics derived from ADC and T2WI images effectively identify PCa in patients with PSA levels of 4–10 ng/mL, significantly outperforming the PI-RADS v2.1 score. 

Source: bmcurol.biomedcentral.com/articles/10.1186/s12894-024-01625-2